Particles, drawing conditional connections?

dtr

Hey, I'm currently playing with 3D particle systems. In the past I have used shiva/vishnu, java code and now gen (tanx for the samples Wes!). I'm looking for a way to draw conditional connecting lines between particles. A basic case would be to connect all particles that are within range x from each other. While I can see this done in java I'm unsure about ways to do it with regular objects and/or gen.

I think I'd need a way of iterating through the particles and build a command list for jit.gl.sketch, or to format a vertex position matrix for jit.gl.mesh.

Any ideas for efficient approaches?

dtr

Here's a solution using js. Still curious about whether there's a way with gen.

var d = 1.0;

// get the patcher object
p = this.patcher;

// get the sketch object
var sketcher = p.getnamed("sketchy");
if (sketcher == null){
post("couldnt find sketch object\n");
} else{
post("bind successful\n");
}

// get the field matrix
var fieldmat = new JitterMatrix("field");

function bang()
{

sketcher.reset();

for(var i = 0; i < 100; i++) {

for(var j = 0; j < 100; j++) {

            var p1 = fieldmat.getcell(i);
            var p2 = fieldmat.getcell(j);
            var distsum = (p1[0]-p2[0])*(p1[0]-p2[0]);
            distsum+= (p1[1]-p2[1])*(p1[1]-p2[1]);
            distsum+= (p1[2]-p2[2])*(p1[2]-p2[2]);
            //post(distsum+"\n");
            if(distsum
                sketcher.linesegment(p1[0],p1[1],p1[2],p2[0],p2[1],p2[2]);

            }
        }
    }
}

function dist(v)
{
d = v;

}

dtr

(Btw, the forum's code tag is malfunctioning there)

dtr

Still looking for more efficient approaches. The js method above, inside my larger system, can only process 40 particles while maintaining 60fps. Pretty bad on a quad-core 3GHz machine.

dtr

The issue should have been clear to me right away: it's the iterating through the particles. Jitter tutorial 46 exemplifies that well. Matrix op or expr in js is far more efficient.

I 'm now working on a basic particle system with one 'attractor'. Not really an attractor cause what I do is draw a line between 2 particles when both are within range x from the attractor and within range x of each other. I can make it a lot more efficient by calculating the distances from particles to the attractor using the op method of tutorial 46, but then I still can't get around some iteration to check the particles I'm left with after the range check. Depending on how many that is I experience fps dips.

Ideas?

...and now gen (tanx for the samples Wes!).

Could you please post the link to the samples? And also thanks for the hints so far in this topic...this is something I plan to do in near future.

just found the patches from Wes...happy days:)